Out-of-Sample Testing for GANs
arXiv:1901.09557
Abstract
We propose a new method to evaluate GANs, namely EvalGAN. EvalGAN relies on a test set to directly measure the reconstruction quality in the original sample space (no auxiliary networks are necessary), and it also computes the (log)likelihood for the reconstructed samples in the test set. Further, EvalGAN is agnostic to the GAN algorithm and the dataset. We decided to test it on three state-of-the-art GANs over the well-known CIFAR-10 and CelebA datasets.
References in corpus (9)
- Adam: A Method for Stochastic Optimization
- Auto-Encoding Variational Bayes
- Explaining and Harnessing Adversarial Examples
- Improved Training of Wasserstein GANs
- Improved Techniques for Training GANs
- Unrolled Generative Adversarial Networks
- Deep Models Under the GAN: Information Leakage from Collaborative Deep Learning
- Learning in Implicit Generative Models
- Generative Visual Manipulation on the Natural Image Manifold